Blocks are ordered chunks of content within a File.

A Block may have raw data, plain text, or both. The type of content is indicated by its mime_type. Blocks can store images, videos, audio clips, or any other chunk of data.

This means that different packages and plugins may choose divide files into blocks using different schemes. Consider a CSV file uploaded to Steamship. The following divisions of this file into blocks are all perfectly fine:

  • Each CSV row is a block of text.

  • Each 10 CSV rows is a block

  • The entire CSV file is one block

Metadata and annotations about the content of the Block added via Tags on the Block .

Creating Blocks#

Blocks may be created when creating a File by passing them in the blocks parameter, or they can be appended to an existing file.

Please see Block.create() and File.append_block().

Read the Block PyDoc spec here.

Making Block Data Public#

If you want the raw data bytes of a Block to be publicly accessible, you can set the parameter public_data = True when calling Block.create(). This is useful if you wish to share a generated image or audio file, or must make the content viewable in a place that cannot retain your Steamship API key. You can also change the value of the public_data flag on an existing Block by calling Block.set_public_data.